|Publication number||US7717850 B2|
|Application number||US 10/997,059|
|Publication date||May 18, 2010|
|Filing date||Nov 24, 2004|
|Priority date||Nov 26, 2003|
|Also published as||CA2545844A1, CA2545846A1, CA2545846C, CN1886095A, CN100566664C, CN101384921A, CN101384921B, CN101849844A, CN101849844B, CN102512200A, EP1691690A1, EP1691690B1, EP1706036A2, EP1706036B1, US8641627, US20050143657, US20050148871, US20100125200, US20100125210, US20100179433, US20140243672, WO2005053540A2, WO2005053540A3, WO2005053541A1|
|Publication number||10997059, 997059, US 7717850 B2, US 7717850B2, US-B2-7717850, US7717850 B2, US7717850B2|
|Inventors||Scott L. Roth, Harold M. Hastings|
|Original Assignee||Imacor Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (43), Non-Patent Citations (6), Referenced by (2), Classifications (26), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims priority to U.S. provisional application No. 60/525,330, filed Nov. 26, 2003.
In the medical field, monitoring heart function impacts critical decisions that relate to patient care. One type of prior art heart monitor is the intravascular/intracardiac ultrasound transducer (such as the Accunav™ transducer). This type of transducer, however, is not well suited for transesophageal echocardiography because the transducer elements are oriented longitudinally instead of transversely, which limits the types of images that can be obtained. A second type of prior art heart monitor is the transesophageal echocardiography (TEE) transducer, which is transversely oriented. However, in order to produce repeatedly usable images, the azimuthal aperture of these transducers must be quite large (e.g., 10-15 mm in diameter for adults), which requires a correspondingly large probe. Because of this large probe, conventional TEE often requires anesthesia, can significantly threaten the airway, and is not well suited for long-term monitoring of the heart.
Transesophageal ultrasound imaging is implemented using a miniature transversely oriented transducer that is preferably small enough to fit in a 7.5 mm diameter probe, and most preferably small enough to fit in a 5 mm diameter probe. Signal processing techniques provide improved depth of penetration, despite the fact that the transducer is so small.
At the other end of the handle 56 is a cable 54, which terminates, at the proximal end of the probe 50, at connector 52. This connector 52 is used to connect the probe 50 to the ultrasound system 200 so that the ultrasound system 200 can operate the probe. Signals for the ultrasound system 200 that drive the transducer 10 travel through the probe 50 via appropriate wiring and any intermediate circuitry (not shown) to drive the transducer 10, and return signals from the transducer 10 similarly travel back through the probe 50 to the ultrasound system 200 where they are ultimately processed into images. The images are then displayed on the monitor 210 in a manner well known to persons skilled in the relevant art.
In the preferred embodiments, the housing 60 has an outer diameter of less than 7.5 mm. The probe contains the ultrasound transducer 10 and connecting wires, and the housing 60 can be passed through the mouth or nose into the esophagus and stomach.
The returned ultrasound signals are processed in the ultrasound system 200 to generate an image of the heart. Preferably, additional signal processing is used to significantly improve image production, as described below.
Unlike conventional TEE systems, the relatively narrow housing used in the preferred embodiments makes it possible to leave the probe in position in the patient for prolonged periods of time.
As best seen in
Other transducer positions may also be used to obtain different views of the heart, typically ranging from the mid-esophagus down to the stomach, allowing the operator to directly visualize most of the relevant cardiac anatomy. For example, the transducer 10 may be positioned in the lower esophagus, so as to obtain the conventional four chamber view. Transducer positioning in the esophagus would typically be done without fully flexing the probe tip, prior to advancing further into the stomach. Within the esophagus, desired views of the heart may be obtained by having the operator use a combination of some or all of the following motions with respect to the probe: advance, withdraw, rotate and slight flex.
For use in adults, the outer diameter of the housing 60 is preferably less than about 7.5 mm, more preferably less than about 6 mm, and is most preferably about 5 mm. This is significantly smaller than conventional TEE probes. This size reduction may reduce or eliminate the need for anesthesia, and may help expand the use of TEE for cardiac monitoring beyond its previous specialized, short-term settings. When a 5 mm housing is used, the housing is narrow enough to pass through the nose of the patient, which advantageously eliminates the danger that the patient will accidentally bite through the probe. Alternatively, it may be passed through the mouth like conventional TEE probes. Note that the 5 mm diameter of the housing is similar, for example, to typical NG (naso-gastric) tubes that are currently successfully used long-term without anesthesia in the same anatomical location. It should therefore be possible to leave the probe in place for an hour, two hours, or even six hours or more.
The housing wall is preferably made of the same materials that are used for conventional TEE probe walls, and can therefore withstand gastric secretions. The wiring in the probe that connects the transducer to the rest of the system may be similar to that of conventional TEE probes (adjusted, of course, for the number of elements). The housing is preferably steerable so that it can be inserted in a relatively straight position, and subsequently bent into the proper position after it enters the stomach. The probe tip may be deflected by various mechanisms including but not limited to steering or pull wires. In alternative embodiments, the probe may use an intrinsic deflecting mechanism such as a preformed element including but not limited to pre-shaped materials. Optionally, the probe (including the transducer housed therein) may be disposable.
For imaging the TGSAV of the LV, the probe tip is preferably ultimately “ante-flexed” (flexed towards the front of the patient) approx. 70-110 degrees. This may be implemented, for example, by building a triggerable ante-flex (e.g., on the order of 70 degrees) into the probe through a combination of a pre-formed element, a device to prevent flexing during insertion and a trigger to release the preformed element from the insertion limit once the probe is in the desired anatomic location. Optionally, a pull-wire may be used for steering to provide the additional 0-40 degrees of flex after the transducer is lowered to the appropriate depth. The triggerable ante-flex component is preferably designed so that it will present little resistance to returning to the unflexed position during probe removal.
Electrical noise may be further reduced using a variety of techniques. For example, in one embodiment, the interface box 203 houses a preamplifier that serves as the first stage in the amplification/processing chain, and separate power supplies are used for the interface box and the main processing unit 201 to reduce electronic noise pass-through. In another embodiment, the interface box 203 houses a preamplifier that serves as the first stage in the amplification/processing chain, and the preamplifier operates on battery power. For both of these embodiments, time gain compensation (TGC) is preferably implemented in that preamplifier. TGC compensates for the fact that the return signals from distant scatterers are weaker than those for nearby scatterers by increasing the gain for signals with longer travel time. TGC may be implemented using conventional techniques that are well known to persons skilled in the relevant art. An example of suitable gain vs. delay characteristics for TGC is shown in
Depth(in cm)=0.077 cm/μs×delay(in μs).
Implementing TGC in the preamplifier facilitates efficient digitization. The preamplifier may also provide amplitude compandoring (a form of compression) to further facilitate efficient digitization. Optionally, the preamplifier's output may be digitized in the interface box, in which case only digital signals would be sent from the interface box to the main processing unit to further reduce electrical noise. These digital signals may even be opto-isolated to eliminate all possible electrical connections in the return path, to reduce electrical noise pass-through further still.
The preferred embodiments described herein provide a good quality image of the TGSAV of the LV from a transducer that is small enough to fit in the narrow housing described above.
As best seen in
The transducer 10 is preferably a phased array transducer made of a stack of N piezo elements L1 . . . LN, an acoustic backing 12, and a matching layer in the front (not shown), in a manner well known to those skilled in the relevant art. As understood by persons skilled in the relevant arts, the elements of phased array transducers can preferably be driven individually and independently, without generating excessive vibration in nearby elements due to acoustic or electrical coupling. In addition, the performance of each element is preferably as uniform as possible to help form a more homogeneous beam. Optionally, apodization may be incorporated into the transducer (i.e., tapering the power driving transducer elements from a maximum at the middle to a minimum near the ends in the azimuthal direction, and similarly tapering the receive gain).
The preferred transducers use the same basic operating principles as conventional TEE transducers to transmit a beam of acoustic energy into the patient and to receive the return signal. However, while the first preferred transducer 10 shown in
TABLE 1 conventional TEE first preferred Feature transducer transducer Size in the transverse 10-15 mm about 4-5 mm (azimuthal) direction Number of elements 64 about 32-40 Size in the elevation direction 2 mm about 4-5 mm Front face aspect ratio about 1:5 about 1:1 (elevation:transverse) Operating frequency 5 MHz about 6-7.2 MHz
A second preferred transducer 10′ is shown in
Size in the transverse
about 4-5 mm
Number of elements
Size in the elevation direction
about 8-10 mm
Front face aspect ratio
about 6-7.2 MHz
In alternative embodiments, the transducer 10 may be built with a size in the elevation direction that lies between the first and second preferred transducers. For example, it may have a size in the elevation direction of about 7.5 mm, and a corresponding elevation:transverse aspect ratio of about 1.5:1.
The transducer 10 preferably has the same transverse orientation (with respect to the axis of the housing 60) as conventional TEE transducers. When the transducer is positioned in the stomach (as shown in
TABLE 3 first second third Parameter example example example number of elements in the transducer 38 36 40 a (azimuthal aperture) 4.75 mm 4.50 mm 5.00 mm b (thickness) 1.25 mm 2.00 mm 2.00 mm c (inner diameter of housing 4.91 mm 4.92 mm 5.39 mm at the transducer) housing wall thickness 0.04 mm 0.04 mm 0.04 mm outer diameter of housing 4.99 mm 5.00 mm 5.47 mm
Referring now to the top view 22 in
The above-describe embodiments assume that the housing is round. However, other shaped housings may also be used to house the transducer, including but not limited to ellipses, ovals, etc. In such cases, references to the diameter of the housing, as used herein, would refer to the diameter of the smallest circle that can circumscribe the housing. To account for such variations in shape, the housing may be specified by its outer perimeter. For example, a 5 mm round housing would have a perimeter of 5π mm (i.e., about 16 mm). When a rectangular transducer is involved, using an oval or elliptical housing can reduce the outer perimeter of the housing as compared to a round housing. For example, an oval that is bounded by a 6 mm×2 mm rectangle with its corners rounded to a radius of 0.5 mm contains a 5 mm×2 mm rectangular region, which can hold the third example transducer in Table 3. Allowing for a 0.04 mm housing wall thickness yields an outer perimeter of 15.4 mm, which is the same outer perimeter as a 4.9 mm diameter circle. The following table gives the outer perimeters that correspond to some of the diameters discussed herein:
Since the characteristics of the last one or two elements at each end of the transducer may differ from the characteristics of the remaining elements (due to differences in their surroundings), the last two elements on each side may be “dummy” elements. In such a case, the number of active elements that are driven and used to receive would be the total number of element (shown in Table 3) minus four. Optionally, the wires to these dummy elements may be omitted, since no signals need to travel to or from the dummy elements. Alternatively, the wires to may be included and the last two elements may be driven, with the receive gain for those elements severely apodized to compensate in part for their position at the ends of the transducer.
Preferably, conventional beam-forming techniques are used to generate and aim a beam of acoustic energy in the desired directions. For example, focusing in the azimuthal direction may be accomplished by phasing (i.e., timing the excitation of individual elements L1 . . . LN in the array, and using appropriate time delays in the returns of individual elements before summing the respective returns into an ultrasound return signal). Focusing in the elevation direction may be accomplished based on the near-field and far-field properties of the sound signal, and will depend upon the physical height of the elements in the elevation direction and optional acoustic lenses.
Resolution adequate to determine LV size and function depends upon a combination of resolution in azimuth, elevation, and axis. This combination is referred to as “spatial resolution” and is illustrated in
The general formula for azimuthal and elevation resolutions is:
where Δθ denotes the beamwidth in radians, λ the wavelength (corresponding to the transducer center frequency) and d the aperture in the given direction (azimuth or elevation). The wavelength λ and aperture d are measured in the same units (e.g., μm).
Axial resolution depends indirectly upon the wavelength λ. Although the inventors are not aware of any specific formula for axial resolution, it is typically on the order of 16-64 times the wavelength. Thus, increasing the center frequency increases all three components of spatial resolution. A center frequency on the order of 5-10 MHz is high enough to provide adequate resolution.
For the first preferred transducer, the elevation aperture is approximately the same as the azimuth aperture. In other words, the front face of the transducer has a elevation:transverse aspect ratio that is approximately 1:1 (i.e., it is approximately square). A square transducer with a width of 4-5 mm in the transverse direction would therefore have an area of approximately 16-25 mm2.
The formulas for azimuthal and elevation resolution are:
where ΔθAZ and ΔθEL are the azimuth and elevation resolutions, respectively, (both measured in radians); and dAZ and dEL are the azimuth and elevation apertures, respectively. These components may be combined into a single equation for overall resolution as a function of area and frequency, as follows:
ΔθOVERALL=1.5×λ2/(d AZ ×d EL)
As explained above, increasing the center frequency results in increased resolution. However, increasing the center frequency also reduces the penetration depth due to frequency-dependent attenuation, which is governed by the approximate formula
where a denotes the one-way attenuation in dB, f is the center frequency in MHz, and r the depth in cm. Thus, one-way frequency dependent attenuation will typically be about 0.5 dB MHz−1 cm−1 and typical round-trip frequency dependent attenuation will typically be about 1 dB MHz−1 cm−1.
The inventors have determined that a transducer center frequency between about 6 and 7.2 MHz provides a good trade-off between resolution and depth of penetration for TEE using a transducer with a 4.75 mm azimuthal aperture. In the embodiments described herein, that range of frequencies can typically provide enough depth of penetration to image the far wall of the left ventricle (in the TGSAV) so that the interior volume of the left ventricle can be computed. (In most subjects, a 12 cm depth of penetration is adequate to image the far wall. For many subjects, a depth of penetration of about 9-10 cm will suffice).
When the transducer elements are spaced at a 125 μm pitch, using a transducer center frequency of 6.16 MHz is particularly advantageous because it corresponds to a wavelength of λ=250 μm. At that wavelength, the elements are spaced at a pitch of 0.5λ, which is sometimes referred to as “half-wavelength pitch”. As is well known to those skilled in the art, a half-wavelength pitch is excellent for eliminating grating lobes while still minimizing element count for a given azimuth aperture. Somewhat larger pitches, e.g. 0.6λ, still work reasonably well in terms of eliminating grating lobes. Thus, for transducers that can operate at a range of center frequencies, acceptable performance can be maintained even if the frequency is increased about 20% (i.e., to the point where the pitch becomes about 0.6λ).
As explained above, the formula for the angular resolution is θ≈λ/d. Referring to the tables above, the first example of the first preferred transducer has a 38 element transducer with a 125 μm pitch, resulting in a 4.75 mm transducer width (d=4750 μm). It is preferably approximately square and operates at a center frequency of 6.16 MHz (λ=250 μm). When those values for d and λ are plugged into the equation for resolution, the result is θ≈0.053 radians, which converts to approximately 3 degrees resolution in both azimuth and elevation.
The increased size of the transducer in the elevation direction helps improve the angular resolution of the system in the elevation direction (as compared to a conventional TEE transducer with a 2 mm elevation). This increased resolution in the elevation direction helps compensate for losses in angular resolution in the azimuth direction caused by shrinking the azimuthal aperture down to about 4-5 mm.
The inventors have noticed that increasing the size of the transducer in the elevation direction further, so that it is larger than the size in the azimuthal direction provides improved performance when imaging the far wall of the heart in the TGSAV. This increase in transducer elevation causes the resolution voxel to shrink in the elevation direction at distances that correspond to the far wall of the LV, which results in increased resolution in the elevation direction. The inventors believe that increasing resolution in this direction is helpful at least in part because the far wall is slanted about the Y axis with respect to the front face of the transducers. (The Y axis is shown in
The inventors have determined that the images of the TGSAV are better when the transducer is more than 1.5 times as large in the elevation direction as the transverse direction, and that the best images of the TGSAV are obtained when the transducer is about two times as large in the elevation direction as the transverse direction, as it is for the second preferred transducer 10′ described above in connection with
Instead of the 90° sector width that is typically used in conventional TEE systems, the preferred embodiment uses a smaller sector width (e.g., 60 degrees). Referring now to
After a beam of ultrasound energy is sent into the patient using the transducer described above, the ultrasound return signal is received, preferably by the same transducer. The transducer converts the ultrasound return signal into an electrical return signal. This process continues as the beam is swept through the imaging sector.
The electrical return signal can be modeled as being an amplitude-modulated signal, with the carrier frequency at the center frequency, and with the modulation being caused in large part from scatterer spacing and other tissue characteristics such as the presence of connective tissue around heart muscle bundles. The electrical return signal is demodulated and digitized (i.e., sampled) to form a demodulated and digitized return signal (DDRS). A variety of conventional techniques that are well known to persons skilled in the relevant arts may be used to form the DDRS. One example is to digitize the electrical return signal and then rectify the result (i.e., take the absolute value) to form a rectified digitized ultrasound return signal. Another example is to rectify the electrical return signal in analog form, and then digitize the result to form the DDRS. Alternative demodulation approaches may also be used to extract the modulation information from the electrical return signal, including but not limited to coherent demodulation, Hilbert transforms, and other demodulation techniques that are well known to persons skilled in the relevant arts.
The samples of each scan line are preferably processed by two different algorithms: one algorithm that analyzes intensity characteristics of the samples, and one algorithm that analyzes frequency characteristics of the samples.
For the first algorithm (i.e., the intensity algorithm) the samples of the scan line is divided into a plurality of pixels, with each pixel containing a plurality of samples. In the
The second algorithm (i.e., the frequency algorithm) analyzes the frequency characteristics of the sample space and determines the spatial frequencies in scatterer spacing. Examples of suitable algorithms are described in U.S. Pat. No. 5,417,215 (hereinafter “the '215 patent”), which is incorporated herein by reference. The article “Spectral Analysis of Demodulated Ultrasound Returns: Detection of Scatterer Periodicity and Application to Tissue Classification” by S. Roth, H. M. Hastings, et al., published in Ultrasonic Imaging 19 (1997) at pp. 266-277, is also incorporated herein by reference.
The frequency algorithm provides a second result for each pixel in the image (i.e., in addition to the result produced by the intensity algorithm). Because most frequency analyzing algorithms provides better results when a larger number of data samples is used, and because each pixel only has a limited number of samples, samples on either side of the pixel in question are preferably combined with the samples in the pixel itself to increase the number of samples. In the illustrated example, each pixel contains 16 samples, but the frequency algorithm for any given pixel operates on 64 samples that are preferably centered in the pixel, as indicated by the boxes labeled “Window for Freq. Algorithm for Pixel k” (where k is an integer from 2-6) that appear below the samples. In this case, for example, all the samples from pixels 2-4 plus half the samples from pixels 1 and 5 would be used to perform the frequency analysis for pixel 3. Of course, other numbers of samples could be used for the frequency analysis instead of 64. Powers of 2, however, are preferable when a fast Fourier transform (FFT) algorithm is used. Optionally, windowing techniques (such as Hamming windows) may be used to weight the samples in the center more heavily than the samples that are near the ends.
In step 1, a Fourier analysis is performed on the samples to determine the power distribution in the various frequency bands at each pixel. The end result of the Fourier analysis of step 1 is a set of amplitude coefficients for each of a plurality of different frequencies, for each of the pixels (i.e., one set of coefficients for the first pixel, a second set of coefficients for the second pixel, etc.). The Fourier analysis may be implemented using any of a variety of algorithms that are well known to persons skilled in the relevant arts (e.g., a conventional FFT algorithm). Alternative embodiments may use other frequency analysis tools to achieve similar results, such as bandpass techniques (preferably integer-based FIR recursive), wavelet techniques, etc. In step 2, the ratio of power in a selected frequency band to the power in the entire spectrum for each pixel is computed. Thus, for each pixel, the following formula applies:
R=E BAND /E TOTAL
Where EBAND is the power in the selected frequency band, ETOTAL is the total power in the portion of the spectrum, and R is the ratio of those two powers. When a Fourier analysis is used, the power in any given band equals the sum of the squares of the amplitudes of the Fourier coefficients within the band. The “selected frequency band” in this step is preferably selected so that changes in the ratio R are correlated to differences in the material that is being imaged (e.g., blood v. muscle). Alternatively, it may be selected so that changes in the ratio R are correlated to differences in S/N ratio, with larger Rs being correlated with signal and smaller Rs being correlated with speckle or electric noise. Optionally, different “selected frequency bands” may be used for near returns and for distant returns. For example, a wider frequency band may be used for signals that correspond to distant structures. In other words, the band selection can be a function of depth.
One suitable set of numeric values that results in a correlation between R and the material being imaged will now be discussed. Consider first the ultrasound return from a single scatter at a depth of r mm. The return from this scatter arrives after a time delay of t μs, given by
t=r/v=r/(0.77 mm/μs)=1.30 r μs,
where the scaling factor of 0.77 mm/μs represents a round trip from the transducer to the scatterer and back (assuming the velocity of sound in tissue is to be 1.54 mm/μs).
The effects of scatterer periodicity upon the spectrum of the demodulated ultrasound return may be calculated in the case of separations large enough so that the ultrasound returns do not overlap (i.e., separations Δr larger than Δr0=0.77 mm/μs×Δt). For example, in the case of an ideal one cycle pulse, a 5 MHz center frequency, and a ideal wide-bandwidth transducer,
Δt=1/fc=1/(5 MHz)=0.200 μs,
Δr0=0.77 mm/μs×0.200 μs=0.154 mm.
The internal structure of cardiac muscle displays variations on this and larger spatial scales. In contrast, scattering from blood is characterized by full-developed speckle, including variations on all, and especially much smaller spatial scales. As a result, low frequencies are indicative of muscle, and high frequencies are indicative of blood. This suggests defining the upper limit of a low frequency band to be less than about 4 MHz, corresponding to a minimal spatial scale ΔrMIN of
ΔrMIN=0.77 mm/μs×1/(4 MHz)=0.77 mm/μs×250 μs=0.193 mm.
The inventors have performed tissue experiments that used a signal digitized at 50 MHz (corresponding to sampling interval of 0.02 μs), and computed the FFT in a 64 point window (corresponding to 64×0.02 μs=1.28 μs, or 0.986 mm). With that size window, the inventors selected a low frequency band that included Fourier frequencies of between 2 and 5 cycles per window (inclusive), which corresponds to frequencies between 2/1.28 MHz=1.56 MHz and 5/1.28 MHz=3.91 MHz.
Using the formula for R set forth above (R=EBAND/ETOTAL) for this low frequency band, the ratio of Fourier power in the low frequency band to the total Fourier power is computed for each pixel. The end result of step 2 in
The inventors have found that, for the parameter values used in this example, R-values of around 0.45 are significantly correlated to the presence of muscle tissue at the pixel of interest, and R-values of around 0.20 are significantly correlated to blood or regions dominated by electronic noise. The remaining part of the algorithm uses this information to improve the image by increasing the intensity of the portions of the image that correspond to muscle and decreasing the intensity of the portions of the image that correspond to blood. Since blood is less reflective than muscle, this difference enhances the contrast between blood and muscle.
The inventors have determined that cardiac ultrasound images are dramatically improved when the intensity of the areas with R-values corresponding to muscle is increased to about 120% of its original value, and when the intensity of the areas with R-values corresponding to blood is decreased to between about 20% and 50% of its original value. Thus, in step 3 of
While most pixels in most images will have R-values that permit the pixel to be classified as being either muscle or blood, in some cases the classification is less clear. For example, pixels that straddle a boundary between muscle and blood have less predictable R values. In addition, although the R-values from blood may average out to 0.20, any given pixel of blood may vary widely from that R-value due to random statistical variations. Accordingly, a monotonic, preferably smooth function may be used to map R to FGF in some embodiments.
Finally, in step 4 of
The actual choice of the Fourier frequency bands, R-values and corresponding FGF values depends upon a variety of factors including but not limited to transducer center frequency, sampling rate, window size and any optional windowing techniques used in signal processing, transducer bandwidth, the width of interrogating pulse, etc. In one embodiment, for example, a transducer center frequency of 7.5 MHz is used, the scan line is digitized at about four times the center frequency (i.e., about 30 MHz), and the distance between the samples is about 0.026 mm.
In alternative embodiments, other normalized (i.e., non-amplitude-dependent measures) may be used instead of dividing EBAND by ETOTAL. For example, the ratio of power in a first frequency band to the power in a second frequency band may be used to compute R, as explained in the '215 patent (e.g., by dividing EBAND1 by EBAND2). In alternative embodiments, two or more Fourier analyses may be performed for each pixel, using a corresponding number of lines of samples, where the center of each line is contained within the pixel. For example, a two line per pixel arrangement in which a first 1D Fourier analyses is implemented along a line of samples in the radial direction, and a second 1D Fourier analyses is implemented along a second line of samples in the tangential direction. The results from those two lines of samples are then merged (e.g., by averaging). In still other embodiments, a 2D Fourier algorithm may be used instead of the 1D algorithms described above.
Ordinarily, the above-described operations are performed on the uncompressed image data. Under certain circumstances, however, it may be possible to perform corresponding operations directly on a compressed version of the image data.
Once the enhanced images have been generated, they may be displayed using conventional hardware. The images may be continuously updated and displayed for the entire time that the probe is in position, so that the physician can visualize the patient's heart in real time. In alternative embodiments, images may be acquired and optionally stored periodically (e.g., by capturing one or more complete heartbeats every two minutes). Optionally, the ability to compare a prior heartbeat to the current heartbeat may be provided by, for example, playing back a stored video clip (or “loop”) of an old heartbeat in one window, and displaying the current image in a second window.
In contrast to conventional extended duration TEE using a transducer with a 10-15 mm azimuthal aperture, which is ordinarily done only under general anesthesia in the closely monitored environment of an room, the smaller diameter of the preferred embodiments described herein permits the preferred embodiments to be used without general anesthesia, and in less closely monitored environments. Optionally, the preferred embodiments may be used with sedation or local anesthesia in place of the general anesthesia that was used with conventional extended duration TEE. It may even be possible to forgo the use of sedation or anesthesia altogether. In such cases, the patient may optionally be medicated with an analgesic.
Optionally, regions of high relevance as detected by the feature gain factor may be highlighted, typically using colorization, while preserving the intensity of the gray-scale image, as explained in the '215 patent. Additional techniques for image enhancing can be found in U.S. Pat. No. 6,932,770, entitled “Method and Apparatus for Ultrasonic Imaging,” which is incorporated herein by reference.
The preferred embodiments described above advantageously permit non-invasive, intermediate and long-term monitoring of cardiac function using a small transducer that fits into a housing approximately 5 mm in diameter, thereby reducing or eliminating the need for anesthesia. The preferred embodiments described above combine a plurality of techniques to produce images that are comparable to or better than images that were conventionally obtained by much larger transducers. The images produced by the preferred embodiments described above are repeatably and reliably usable for monitoring heart function, with adequate penetration depth to see the far wall of the left ventricle (10-12 cm) and adequate resolution to determine LV size and function from an image of the endocardial wall in real time, despite the use of a smaller transducer. Thus, in contrast to prior art systems which provide a depth of penetration that is less than 15 times the azimuthal aperture of the transducer (e.g., obtaining 10 cm penetration using a 10 mm transducer) the preferred embodiments can provide penetration that is greater than 15 times the azimuthal aperture of the transducer, or even greater than 20 times the azimuthal aperture of the transducer (e.g., obtaining 10 cm penetration using a 4.75 mm transducer).
The preferred embodiments described above use a probe that is much narrower than conventional TEE probes, and may be used to monitor heart function over an extended period of time and to obtain an understanding of the patients' hemodynamic status. Such information may be useful in choosing treatments and improving outcome in many situations (including but not limited to critical medical problems such as hypotension, pulmonary edema and heart failure).
The above-described embodiments permit direct visualization of cardiac function, which permits evaluation of a patient's hemodynamic status including intravascular volume (normal, low or high), cardiac contractility (how well the left ventricle pumps), cardiac ischemia (inadequacy of blood flow to the heart muscle) and cardiac tamponade (fluid in the pericardial sac limiting heart function). For example, information about intravascular volume status can be derived from directly visualizing the size of the left ventricle and monitoring changes in size with treatments over time. Information about contractility can be obtained by directly visualizing the contraction (pumping) of the left ventricle, either using qualitative visual estimates or quantitatively. Information about ischemia is available during direct visualization of the left ventricle since ischemia results in abnormal motion of the walls of the left ventricle (wall motion abnormality). Information about possible cardiac tamponade or pericardial effusion (fluid in the heart sac) is available when using ultrasound to directly visualize the heart.
The narrowness of the probe may enable the above-described embodiments to provide this information for longer periods of time, outside the operating room, and/or without anesthesia. The above-described embodiments also lend themselves to use in settings where interventional cardiac procedures are performed such as the cardiac catheterization and electrophysiology laboratories, both for monitoring the effects of physicians' interventions on cardiac and hemodynamic function and for guiding the placement of devices. For example, they may be used to help the physician correctly place the pacing leads to achieve the desired result. The above-described embodiments may also be used in non-cardiac applications in which a narrower probe is needed or beneficial.
The above-described embodiments are not limited to ultrasound imaging modes, and may be used in alternative ultrasound modes (e.g., pulsed wave Doppler, continuous wave Doppler, and color flow imaging Doppler modes). These alternative modes may be performed using the same transducer as the above-described imaging modes and may yield information which can be combined with images, optionally in real-time. For example, color flow Doppler information may be obtained during imaging of the mitral valve (between the left atrium and left ventricle) while maintaining transducer position in the mid to lower esophagus. Such an application would permit evaluation of leakage of the mitral valve (mitral regurgitation or insufficiency).
If desired, the preferred embodiments described above may be scaled down for neonatal or pediatric use. In such cases, a transducer that is between about 2.5 and 4 mm in the azimuthal direction is preferable, with the elevation dimension scaled down proportionally. Because less depth of penetration is required for neonatal and pediatric patients, the operating frequency may be increased. This makes λ smaller, which permits the use of a smaller transducer element spacing (pitch), and a correspondingly larger number of elements per mm in the transducer. When such a transducer is combined with the above-described techniques, the performance should meet or surpass the performance of conventional 7.5 mm TEE probes for neonatal and pediatric uses.
The embodiments described herein may also be used in non-cardiac applications. For example, the probe could be inserted into the esophagus to monitor the esophagus itself, lymph nodes, lungs, the aorta, or other anatomy of the patient. Alternatively, the probe could be inserted into another orifice (or even an incision) to monitor other portions of a patient's anatomy.
If desired, the center frequency may be lowered (e.g., down to about 4.5 MHz) to provide additional depth of penetration when needed (e.g., for very large patients). Although this will also reduce the resolution, the result may be acceptable when very large structures are being imaged. Alternatively, the transducer size and housing diameter may be scaled up in size (e.g., to about 7 mm) if the reduced resolution results in unusable images.
Numerous alternative and optional features may be substituted and added to the above-described embodiment. One optional feature is digital beamforming using significant oversampling. For example, if the transducer is operated at 7 MHz, and the return is digitized at 30× frequency, 30×7 MHz=210 MHz digitization would be required. That data could then be downsampled by a factor of five to reduce the number of data points to a 42 MHz sample. Such downsampling would reduce the noise floor due to front-end noise by a factor of √5, (i.e., over 2 bits in power). Similarly, downsampling by a factor of 7 would reduce the noise floor by a factor of √7.
Alternative embodiments of the invention may use fewer techniques and/or implement those techniques to a lesser extent, and still maintain the ability to produce an acceptable image. For example, depending on the other components in the system, it may be possible to obtain an acceptable image using a 75° sector width, or even using a 90° sector width. It may also be possible to obtain an acceptable image using a transducer with an elevation:transverse aspect ratio of about 2:3 in place of the preferred 1:1 or 2:1 aspect ratios. Another alternative would be to use some or all of the above-described techniques with a transducer that is slightly larger than the preferred embodiments described above, yet still smaller than conventional 10 mm TEE transducer. Numerous other modifications to the above-described embodiments will be apparent to those skilled in the art, and are also included within the purview of the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4646748||Feb 19, 1985||Mar 3, 1987||Terumo Kabushiki Kaisha||Ultrasonic measurement method, and apparatus therefor|
|US4771470||Nov 14, 1985||Sep 13, 1988||University Of Florida||Noise reduction method and apparatus for medical ultrasound|
|US4817015||Nov 18, 1985||Mar 28, 1989||The United States Government As Represented By The Secretary Of The Health And Human Services||High speed texture discriminator for ultrasonic imaging|
|US4977898||May 30, 1989||Dec 18, 1990||Hoffrel Instruments, Inc.||Miniaturized encapsulated ultrasonic transducer|
|US4982339||Jan 18, 1989||Jan 1, 1991||The United States Of America As Represented By Department Of Health And Human Service||High speed texture discriminator for ultrasonic imaging|
|US5111823||Apr 16, 1990||May 12, 1992||National Fertility Institute||Apparatus and method for generating echographic images|
|US5224483 *||May 19, 1992||Jul 6, 1993||Hewlett-Packard Company||Adaptive contrast enhancement for scanned ultrasonic image|
|US5291893||Oct 9, 1992||Mar 8, 1994||Acoustic Imaging Technologies Corporation||Endo-luminal ultrasonic instrument and method for its use|
|US5325860||Nov 8, 1991||Jul 5, 1994||Mayo Foundation For Medical Education And Research||Ultrasonic and interventional catheter and method|
|US5390676||Dec 8, 1993||Feb 21, 1995||Hitachi, Ltd.||Ultrasonic flowmeter|
|US5409007||Nov 26, 1993||Apr 25, 1995||General Electric Company||Filter to reduce speckle artifact in ultrasound imaging|
|US5417215||Feb 4, 1994||May 23, 1995||Long Island Jewish Medical Center||Method of tissue characterization by ultrasound|
|US5497777||Sep 23, 1994||Mar 12, 1996||General Electric Company||Speckle noise filtering in ultrasound imaging|
|US5520185 *||Nov 15, 1994||May 28, 1996||Cardiovascular Imaging Systems, Inc.||Method for recognition and reduction of blood speckle in blood vessel imaging system|
|US5619998||Mar 11, 1996||Apr 15, 1997||General Electric Company||Enhanced method for reducing ultrasound speckle noise using wavelet transform|
|US5754676 *||Apr 7, 1995||May 19, 1998||Olympus Optical Co., Ltd.||Image classification apparatus|
|US5757727 *||Feb 19, 1997||May 26, 1998||Acuson Corporation||Two-dimensional acoustic array and method for the manufacture thereof|
|US5839441||Jun 3, 1996||Nov 24, 1998||The Trustees Of The University Of Pennsylvania||Marking tumors and solid objects in the body with ultrasound|
|US5957846 *||Nov 17, 1997||Sep 28, 1999||Teratech Corporation||Portable ultrasound imaging system|
|US5961464 *||Sep 16, 1998||Oct 5, 1999||Hewlett-Packard Company||Ultrasound contrast agent detection using spectral analysis from acoustic scan lines|
|US6036645 *||Apr 3, 1998||Mar 14, 2000||Transonic Systems, Inc.||Ultrasonic probe|
|US6059731 *||Aug 19, 1998||May 9, 2000||Mayo Foundation For Medical Education And Research||Simultaneous side-and-end viewing underfluid catheter|
|US6080107 *||Jan 26, 1999||Jun 27, 2000||Hewlett-Packard Company||Methods for the use of contrast agents in ultrasonic imaging|
|US6181810||Jul 30, 1998||Jan 30, 2001||Scimed Life Systems, Inc.||Method and apparatus for spatial and temporal filtering of intravascular ultrasonic image data|
|US6264609||Sep 15, 1999||Jul 24, 2001||Wake Forest University||Ultrasound apparatus and method for tissue characterization|
|US6359637||Feb 5, 1999||Mar 19, 2002||International Busines Machines Corp.||Drill-down apparatus for display of tree-based hierarchies and method therefor|
|US6383139||Oct 22, 1999||May 7, 2002||Sonosite, Inc.||Ultrasonic signal processor for power doppler imaging in a hand held ultrasonic diagnostic instrument|
|US6512854||May 7, 1999||Jan 28, 2003||Koninklijke Philips Electronics N.V.||Adaptive control and signal enhancement of an ultrasound display|
|US6514202||Apr 23, 2001||Feb 4, 2003||Scimed Life Systems, Inc.||Systems and methods for evaluating objects within an ultrasound image|
|US6533728 *||Nov 20, 2001||Mar 18, 2003||Mayo Foundation For Medical Education And Research||Method and apparatus for recovery and parametric display of contrast agents in ultrasound imaging|
|US6572547||Jul 31, 2001||Jun 3, 2003||Koninklijke Philips Electronics N.V.||Transesophageal and transnasal, transesophageal ultrasound imaging systems|
|US6579238 *||Feb 23, 2001||Jun 17, 2003||Acuson Corporation||Medical ultrasonic imaging system with adaptive multi-dimensional back-end mapping|
|US6679844 *||Jun 20, 2002||Jan 20, 2004||Acuson Corporation||Automatic gain compensation for multiple mode or contrast agent imaging|
|US6932770 *||Aug 4, 2003||Aug 23, 2005||Prisma Medical Technologies Llc||Method and apparatus for ultrasonic imaging|
|US20010029336||Apr 20, 2001||Oct 11, 2001||Tat-Jin Teo||Methods and apparatus for blood speckle detection in an intravascular ultrasound imaging system|
|US20010031924||Mar 1, 2001||Oct 18, 2001||Seward James B.||Small ultrasound transducers|
|US20010055025 *||Jan 4, 2001||Dec 27, 2001||Sun Microsystems, Inc.||Graphics system configured to filter samples using a variable support filter|
|US20030092994 *||Dec 13, 2002||May 15, 2003||Miller David G.||Ultrasonic tee probe with two dimensional array transducer|
|US20030231790 *||May 2, 2003||Dec 18, 2003||Bottema Murk Jan||Method and system for computer aided detection of cancer|
|USRE33672||Apr 29, 1987||Aug 27, 1991||Fujitsu Limited||Method for measuring characteristics of living tissue by ultrasonic waves|
|EP0248623A2||Jun 1, 1987||Dec 9, 1987||Washington University||Ultrasound scanning apparatus|
|EP0749723A1||Jun 23, 1995||Dec 27, 1996||Arno Schnorrenberg Chirurgiemechanik||Intestinal ultrasound probe for trans-intestinal diagnosis in birds, reptiles and/or small mammals|
|WO2001065538A2||Mar 1, 2001||Sep 7, 2001||Mayo Foundation||Small ultrasound transducers|
|1||Jérôme Poguet et al., "Phased Array Technology: Concepts, Probes and Applications", NDT.net, (2002), vol. 7, No. 05.|
|2||K. K. Djoa et al., "Two Decades of Transesophageal Phased Array Probes", Ultrasound in Med. & Biol., vol. 22, No. 1., (1996).|
|3||Paul A. Heindenreich, "Transesophageal Echocardiography (TEE) in the Critical Care Patient", Transesophageal Echocardiography, vol. 18, No. 4, (2000), pp. 789-805.|
|4||Roy W. Martin et al., "Design characteristics for intravascular ultrasonic catheters," International Journal of Cardiac Imaging 4:201-216, (1989).|
|5||Scott L. Roth et al., "Spectral Analysis of Demodulated Ultrasound Returns: Detection of Scatterer Periodicity and Application to Tissue Classification", Ultrasonic Imaging 19. (1997).|
|6||Search Report Dated Jul. 11, 2008 from EP 04 75 6114.7.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|WO2012141913A1||Mar 29, 2012||Oct 18, 2012||Imacor, Inc.||Ultrasound guided positioning of cardiac replacement valves|
|WO2012141914A1||Mar 29, 2012||Oct 18, 2012||Imacor, Inc.||Ultrasound guided positioning of cardiac replacement valves with 3d visualization|
|U.S. Classification||600/437, 600/449, 600/443|
|International Classification||G01S15/89, G01S7/521, A61B8/00, G01S7/52, A61B8/12|
|Cooperative Classification||G01S15/8915, G01S7/52036, G01S15/8934, G01S15/8977, G01S7/52033, G01S15/8925, A61B8/0883, A61B8/445, G01S7/52079, A61B8/12, A61B8/4488|
|European Classification||A61B8/44R2, A61B8/12, G01S15/89D1C4, G01S7/52S13, G01S15/89D6, G01S15/89D2, G01S7/52S2F|
|Dec 13, 2010||AS||Assignment|
Effective date: 20101210
Owner name: WFD VENTURES LLC, AS AGENT, NEW YORK
Free format text: SECURITY AGREEMENT;ASSIGNOR:IMACOR, INC.;REEL/FRAME:025493/0201
|Jun 27, 2011||AS||Assignment|
Effective date: 20060605
Free format text: CHANGE OF NAME;ASSIGNOR:PRISMA MEDICAL TECHNOLOGIES LLC;REEL/FRAME:026509/0653
Owner name: IMACOR LLC, NEW YORK
|Jun 28, 2011||AS||Assignment|
Owner name: IMACOR INC., NEW YORK
Effective date: 20090206
Free format text: MERGER;ASSIGNOR:IMACOR LLC;REEL/FRAME:026515/0200
|Aug 29, 2011||AS||Assignment|
Owner name: IMACOR, INC., NEW YORK
Free format text: RELEASE BY SECURED PARTY;ASSIGNOR:WFD VENTURES LLC, AS AGENT;REEL/FRAME:026820/0411
Effective date: 20110829
|Jun 27, 2013||FPAY||Fee payment|
Year of fee payment: 4
|Apr 1, 2014||AS||Assignment|
Effective date: 20140328
Owner name: FRIEDMAN, VALERIE, MASSACHUSETTS
Free format text: SECURITY INTEREST;ASSIGNOR:IMACOR INC.;REEL/FRAME:032572/0335